This document describes the development of an intelligent algorithm for power system fault detection and stability studies. It was developed by a team of students under the guidance of faculty at K.L.E. Dr. M.S. SHESHGIRI College of Engineering and Technology. The document introduces SCADA systems and their limitations, the need for wide area monitoring systems (WAMS) to address these limitations, and the components and objectives of WAMS, which include phasor measurement units and data concentrators. Machine learning algorithms are necessary due to the complex, extensive data handled by WAMS. The conclusion discusses benefits of WAMS such as real-time monitoring and post-disturbance analysis to predict system stability.
This document describes the development of an intelligent algorithm for power system fault detection and stability studies. It was developed by a team of students under the guidance of faculty at K.L.E. Dr. M.S. SHESHGIRI College of Engineering and Technology. The document introduces SCADA systems and their limitations, the need for wide area monitoring systems (WAMS) to address these limitations, and the components and objectives of WAMS, which include phasor measurement units and data concentrators. Machine learning algorithms are necessary due to the complex, extensive data handled by WAMS. The conclusion discusses benefits of WAMS such as real-time monitoring and post-disturbance analysis to predict system stability.
This document describes the development of an intelligent algorithm for power system fault detection and stability studies. It was developed by a team of students under the guidance of faculty at K.L.E. Dr. M.S. SHESHGIRI College of Engineering and Technology. The document introduces SCADA systems and their limitations, the need for wide area monitoring systems (WAMS) to address these limitations, and the components and objectives of WAMS, which include phasor measurement units and data concentrators. Machine learning algorithms are necessary due to the complex, extensive data handled by WAMS. The conclusion discusses benefits of WAMS such as real-time monitoring and post-disturbance analysis to predict system stability.
TECHNOLOGY DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING BELAGAVI-590008
Development Of Intelligent Algorithm for
Power System Fault Detection And Stability Studies Under the Guidance of Dr. Arunkumar Patil Dr. Girish(Executive Engg. HESCOM) Prof. Shashidhar H(CSE) Team members: 1.Omkar.S. Chitnis 2.Vipul Choramule 3.Sushma Mirje 4.Manjula Pidai Contents 1)Introduction 2)What is SCADA and limitations 3)Need of WAMS? 4)What is WAMS? 5)Components of WAMS 6)Objective 7)Methodology 8)Conclusion 9)References Power system Introduction Monitoring Network Supervisory Control And Data Acquisition System(SCADA) SCADA is used for monitoring, analyzing and controlling action Limitations in SCADA 1) Does not provide dynamic behavior of system. 2) Measurements are not time synchronized. 3) Measurements obtained at slower rate of (1-sample/5-10sec). 4) Only gives magnitude measurement, does not provide phase angle. 5) Based on local area monitoring system. Need of WAMS In order to avoid regional blackout that occurred in India on July 30 & 31 of 2012. When constant monitoring applications are available immediate action can be taken if some failures are detected. It allow the networks to operate closer to its capacity while maintaining system security. What is WAMS? Wide Area Monitoring Systems (WAMS) are based on the new data acquisition technology of phasor measurement and allow monitoring transmission system conditions over large areas in view of detecting and further counter acting of grid instabilities instant. Two major functions: 1) Obtaining the data. 2) Extracting value from it. The measured data are used for further development, methodology and verification of synchrophasor based functions. Components of WAMS 1)Phasor Measurement Unit(PMU) It is the devices which use synchronization signal from the global positioning system(GPS) satellite and provide the phasor voltage and currents measured at a given substation. A phasor is a complex number that represents both magnitude and phase angle of the sine waves found in electricity. PMU can measure typical rate of 25 samples per second.
2)Phasor Data Concentrator(PDC)
It is a PMU data collecting device that synchronizes the measurements taken at every instant of time. It also monitors power quality indices such as voltage sag, frequency, active and reactive power. Bad data rejection, error checking and to create a coherent record of simultaneously recorded data. Why WAMS? 1)Real time determination of system parameters using PMU. 2)Early indication of system deterioration with high speed. 3)Accurate assessment and better decision making while emergencies. 4)Reduced system outage time and wide area monitoring. Necessity behind of Machine Learning Algorithm The iterative aspect of machine learning is important because as input samples are exposed to new data, they are able to independently adapt. Increase in the computational time period and accuracy due to extensive and vast system data handling. Complex, versatile and large amount of information which is used in calculation ,diagnosis and learning. Techniques For Implementation Classification REGRESSION CLUSTERING
Support vector Linear Regression Hierarchical
machine Discriminant Gaussian Mixture Ensemble Analysis methods Navie Bayes Neural Networks Decision Trees Algorithms Nearest Neighbor K-Means,Fuzzy C- Neural Networks Means Neural Networks Hidden Markov SVR,GPR Model Conclusion PMU Measures 25 samples per cycle of waveform. Rate of data streaming is 20 to 100ms. Real time monitoring of power system. Post disturbance analysis of power system by which stability of system is predicted. Phasor angle and magnitude measurement. Elimination of time skew. References 1)A. G. Phadke, “Synchronized phasor measurements in power systems,” IEEE Comput. Appl. Power, vol. 6, no. 2, pp 1-15, Apr. 1993. 2) Aneesh Rajeev, Angel T.S and Faraz Zafar Khan, "Optimally Placed PMU's for Fault Detection in Distribution Network," in Proc. IEEE 3) M. Pignati, L. Zanni, P. Romano, R. Cherkaoui, and M. Paolone, “Fault detection and faulted line identification in active distribution networks using synchrophasors-based real- time state estimation,” IEEE Trans. on Power Delivery, vol. 32, no. 1, pp. 381–392, Feb. 2017. 4)www.posoco.ac.in Thank You